ADL: Docking inside recptor (David Crosby)

Jeffrey Dyason j.dyason at
Mon Mar 17 14:40:50 PDT 2008

Have a look at your starting protein crystal structure and you will 
generally see that there are Waters located in these cavities. To stop 
autodock putting the ligand there, include the waters in the calculation 
when setting up your grid for your protein.

Bye Jeff.
Jeff Dyason                                    e-mail: 
J.Dyason at
Institute for Glycomics
Griffith University                          (Gold Coast Campus)  
PMB 50 Gold Coast Mail Centre       Tel: +61 7 5552 7028
Queensland 9726 Australia              Fax: +61 7 5552 8098

"Antonio Cavazzuti" <acavazzuti at> 
Sent by: autodock-bounces at
18/03/08 02:51 AM
Please respond to
autodock at

autodock at

ADL: Docking inside recptor  (David Crosby)

Hi David, 
did you solved the problem concerning the docking inside the receptor?
(problem reported in autodock Digest, Vol 48, Issue 17)
If yes, can you suggest me how to avoid this problem?
Thank you!

> 1)       I've noticed often when I attempt to dock a ligand I'll get 
> a handful (perhaps 5 in 25-50) of dockings where the ligand is 
> docked quite literally inside the macromolecule resulting in some 
> astronomical dG value. Is there some setting I'm missing when I set 
> up my grids or docking parameters to tell Autodock to avoid 
> nonsensical dockings?  I see you can specify an upper energy limit 
> under the docking parameters menu, but can I assign a lower limit to 
> avoid these wild results?  They throw off the clustering data.  I'm 
> attempting to dock a somewhat small molecule with 4-6 torsions onto 
> a relatively small receptor (~6kd,)  I'm trying to keep my initial 
> dockings unbiased, so my grid envelops the entire receptor. I've 
> left plenty of space for the ligand to rotate freely in all 
> configurations, so why does my ligand often end up inside my 
> receptor instead of on the surface?  I understand autodock first 
> introduce 50 randomly placed/oriented ligands inside the grid volume 
> and make calculations from there, but why would it consider 
> molecules making obviously high-energy (or low) contacts with other 
> atoms as preferable dockings?
--- ADL: AutoDock List  --- ---

More information about the autodock mailing list